Cells of malignant breast tumors exhibit centrosome defects including an excess number of centrioles, increased microtubule nucleation capacity, and inappropriate phosphorylation of centrosomal proteins, a condition termed 'centrosome amplification Centrosome amplification leads to multipolar mitosis and consequent chromosomal instability, and therefore, is one mechanism by which aneuploidy and phenotypic variability arise in the development of cancer. We propose that centrosome amplification is an early event in the development of breast cancer, and that amplified centrosomes may arise through one of several alternative mechanisms. We propose to characterize the origin of centrosome amplification during the development of breast tumors. We will test the hypothesis that ER and growth factor signaling are mechanistically coupled to centriole separation and centrosome duplication. And finally, we will experimentally disrupt cell cycle progression in normal breast and tumor-derived cell lines to test the hypothesis that the G 1/S and G2/M cell cycle checkpoints are mechanistically coupled to centrosome duplication and that this linkage becomes uncoupled during mammary tumorigenesis. The study of centrosome behavior is of fundamental importance to our understanding of the origin of malignant tumors and may reveal new targets for intervention or prevention of the development of breast cancer.